← 26edo 27edo 28edo →
Prime factorization 33
Step size 44.4444 ¢ 
Fifth 16\27 (711.111 ¢)
Semitones (A1:m2) 4:1 (177.8 ¢ : 44.44 ¢)
Consistency limit 9
Distinct consistency limit 7

27 equal divisions of the octave (27edo), or 27(-tone) equal temperament (27tet, 27et) when viewed from a regular temperament perspective, is the tuning system derived by dividing the octave into 27 equally large steps. Each step represents a frequency ratio of the 27th root of 2, or 44.4 cents.

Theory

If octaves are kept pure, 27edo divides the octave in 27 equal parts each exactly 44.444… cents in size. However, 27 is a prime candidate for octave shrinking, and a step size of 44.3 to 44.35 cents would be reasonable. The reason for this is that 27edo tunes harmonics 3, 5, and 7 sharply. The optimal step size for octave-shrinking in the 7-limit is 44.3071 cents, which rougly corresponds to 27.0837-edo. Furthermore, 27edo's local zeta peak is at 27.086614-edo, which corresponds to a step size of 44.3023 cents.

Assuming however pure octaves, 27 has a fifth sharp by slightly more than nine cents and a 7/4 sharp by slightly less, and the same 400 cent major third as 12edo, sharp 13+2/3 cents. The result is that 6/5, 7/5 and especially 7/6 are all tuned more accurately than this. It can be considered the superpythagorean counterpart of 19edo, as its 5th is audibly indistinguishable from 1/3 septimal comma superpyth in the same way that 19edo is audibly indistinguishable from 1/3 syntonic comma meantone, resulting in three of them reaching a near perfect minor third/major sixth in both.

27edo, with its 400 cent major third, tempers out the diesis of 128/125, and also the septimal comma, 64/63 (and hence 126/125 also). These it shares with 12edo, making some relationships familiar, and as a consequence they both support augene temperament. It shares with 22edo tempering out the allegedly Bohlen-Pierce comma 245/243 as well as 64/63, so that they both support the superpyth temperament, with quite sharp "superpythagorean" fifths giving a sharp 9/7 in place of meantone's 5/4.

Though the 7-limit tuning of 27edo is not highly accurate, it nonetheless is the smallest equal division to represent the 7-odd-limit both consistently and distinctly – that is, everything in the 7-odd-limit diamond is uniquely represented by a certain number of steps of 27edo. It also represents the 13th harmonic very well, and performs quite decently as a 2.3.5.7.13 temperament. It also approximates 19/10, 19/12, and 19/14, so 0-7-13-25 does quite well as a 10:12:14:19; the major seventh 25\27 is less than a cent off from 19/10. Octave-inverted, these also form a quite convincing approximation of the main Bohlen-Pierce triad, 3:5:7, making it the smallest edo that can simulate tritave harmony, although it rapidly becomes quite rough if extended to the 9 and above, unlike a true tritave based system.

Its step, as well as the octave-inverted and octave-equivalent versions of it, holds the distinction for having around the highest harmonic entropy possible and thus is, in theory, most dissonant, assuming the relatively common values of a = 2 and s = 1%. This property is shared with all edos between around 24 and 30. Intervals smaller than this tend to be perceived as unison and are more consonant as a result; intervals larger than this have less "tension" and thus are also more consonant.

The 27-note system or one similar like a well temperament can be notated very easily, by a variation on the quartertone accidentals. In this case a sharp raises a note by 4 edosteps, just one edostep beneath the following nominal (for example C to C# describes the approximate 10/9 and 11/10 interval) and the flat conversely lowers: these are augmented unisons and diminished unisons. Just so, one finds that an accidental can be divided in half, and this fill the remaining places without need for double sharps and double flats. Enharmonically then, E double flat means C half sharp. In other words, the resemblance to quarter tone notation differs in enharmonic divergence. The notes from C to D are C, D flat, C half-sharp, D half-flat, C sharp, D. Unfortunately, some ascending intervals appear to be descending on the staff. Furthermore, the 3rd of a 4:5:6 or 10:12:15 chord must be notated as either a 2nd or a 4th. The composer can decide for him/herself which addidional accidental pair is necessary if they will need redundancy to remedy these problems, and to keep the chromatic pitches within a compass on paper relative to the natural names (C, D, E etc.). Otherwise it is simple enough, and the same tendency for A# to be higher than Bb is not only familiar, though here very exaggerated, to those working with the Pythagorean scale, but also to many classically trained violinists.

Odd harmonics

Approximation of odd harmonics in 27edo
Harmonic 3 5 7 9 11 13 15 17 19 21 23
Error Absolute (¢) +9.2 +13.7 +9.0 +18.3 -18.0 +3.9 -21.6 -16.1 +13.6 +18.1 -6.1
Relative (%) +20.6 +30.8 +20.1 +41.2 -40.5 +8.8 -48.6 -36.1 +30.6 +40.7 -13.6
Steps
(reduced)
43
(16)
63
(9)
76
(22)
86
(5)
93
(12)
100
(19)
105
(24)
110
(2)
115
(7)
119
(11)
122
(14)

Intervals

# Cents Approximate Ratios* Ups and Downs Notation 6L 1s notation Solfege
0 0.00 1/1 P1 perfect unison D perfect unison C do
1 44.44 28/27, 36/35, 39/38, 49/48, 50/49, 81/80 ^1, m2 up unison, minor 2nd ^D, Eb aug 1sn, double-dim 2nd C#, Dbbb di
2 88.89 16/15, 21/20, 25/24, 19/18, 20/19 ^^1, ^m2 dup unison, upminor 2nd ^^D, ^Eb double-aug 1sn, dim 2nd Cx, Dbb ra
3 133.33 15/14, 14/13, 13/12 vA1, ~2 downaug 1sn, mid 2nd vD#, vvE triple-aug 1sn, minor 2nd Cx#, Db ru
4 177.78 10/9 A1, vM2 aug 1sn, downmajor 2nd D#, vE major 2nd D reh
5 222.22 8/7, 9/8 M2 major 2nd E aug 2nd, double-dim 3rd D#, Ebbb re
6 266.67 7/6 m3 minor 3rd F double-aug 2nd, dim 3rd Dx, Ebb ma
7 311.11 6/5, 19/16 ^m3 upminor 3rd Gb triple-aug 2nd, minor 3rd Dx#, Eb me
8 355.56 16/13 ~3 mid 3rd ^Gb E major 3rd E mu
9 400.00 5/4, 24/19 vM3 downmajor 3rd vF# aug 3rd, double-dim 4th E#, Fbbb mi
10 444.44 9/7, 13/10 M3 major 3rd F# double-aug 3rd, dim 4th Ex, Fbb mo
11 488.89 4/3 P4 perfect 4th G triple-aug 3rd, minor 4th Ex#, Fb fa
12 533.33 27/20, 48/35, 19/14, 26/19 ^4 up 4th Ab major 4th F fih
13 577.78 7/5, 18/13 ~4, ^d5 mid 4th, updim 5th ^^G, ^Ab aug 4th, double-dim 5th F#, Gbbb fi
14 622.22 10/7, 13/9 vA4, ~5 downaug 4th, mid 5th vG#, vvA double-aug 4th, dim 5th Fx, Gbb se
15 666.67 40/27, 35/24, 19/13, 28/19 v5 down fifth G# triple-aug 4th, minor 5th Fx#, Gb sih
16 711.11 3/2 P5 perfect 5th A major 5th G so/sol
17 755.56 14/9, 20/13 m6 minor 6th Bb aug 5th, double-dim 6th G#, Abbb lo
18 800.00 8/5, 19/12 ^m6 upminor 6th ^Bb double-aug 5th, dim 6th Gx, Abb le
19 844.44 13/8 ~6 mid 6th vA# triple-aug 5th, minor 6th Gx#, Ab lu
20 888.89 5/3, 32/19 vM6 downmajor 6th A# major 6th A la
21 933.33 12/7 M6 major 6th B aug 6th, double-dim 7th A#, Bbbb li
22 977.78 7/4, 16/9 m7 minor 7th C double-aug 6th, dim 7th Ax, Bbb ta
23 1022.22 9/5 ^m7 upminor 7th Db triple-aug 6th, minor 7th Ax#, Bb te
24 1066.67 28/15, 13/7, 24/13 ~7 mid 7th ^Db major 7th B tu
25 1111.11 15/8, 40/21, 48/25, 19/10, 36/19 vM7 downmajor 7th vC# aug 7th, double-dim 8ve B#, Cbb ti
26 1155.56 27/14, 35/18, 96/49, 49/25, 160/81 M7 major 7th C# double-aug 7th, dim 8ve Bx, Cb da
27 1200.00 2/1 P8 8ve D 8ve C do

* based on treating 27edo as a 2.3.5.7.13.19 subgroup temperament; other approaches are possible.

Combining ups and downs notation with color notation, qualities can be loosely associated with colors:

quality color name monzo format examples
minor zo {a, b, 0, 1} 7/6, 7/4
fourthward wa {a, b}, b < -1 32/27, 16/9
upminor gu {a, b, -1} 6/5, 9/5
mid tho {a, b, 0, 0, 0, 1} 13/12, 13/8
thu {a, b, 0, 0, 0, -1} 16/13, 24/13
downmajor yo {a, b, 1} 5/4, 5/3
major fifthward wa {a, b}, b > 1 9/8, 27/16
ru {a, b, 0, -1} 9/7, 12/7

All 27edo chords can be named using ups and downs. Alterations are always enclosed in parentheses, additions never are. An up or down after the chord root affects the 3rd, 6th, 7th, and/or the 11th (every other note of a stacked-3rds chord 6-1-3-5-7-9-11-13). Here are the zo, gu, ilo, yo and ru triads:

color of the 3rd JI chord notes as edosteps notes of C chord written name spoken name
zo 6:7:9 0-6-16 C Eb G Cm C minor
gu 10:12:15 0-7-16 C ^Eb G C^m C upminor
ilo 18:22:27 0-8-16 C vvE G C~ C mid
yo 4:5:6 0-9-16 C vE G Cv C downmajor or C down
ru 14:18:21 0-10-16 C E G C C major or C

For a more complete list, see Ups and Downs Notation #Chords and Chord Progressions. See also the 22edo page.

JI approximation

15-odd-limit interval mappings

The following table shows how 15-odd-limit intervals are represented in 27edo. Prime harmonics are in bold; inconsistent intervals are in italic.

15-odd-limit intervals by direct approximation (even if inconsistent)
Interval, complement Error (abs, ¢) Error (rel, %)
7/6, 12/7 0.204 0.5
15/11, 22/15 3.617 8.1
13/8, 16/13 3.917 8.8
5/3, 6/5 4.530 10.2
9/5, 10/9 4.626 10.4
7/5, 10/7 4.734 10.7
13/7, 14/13 5.035 11.3
13/12, 24/13 5.239 11.8
11/9, 18/11 8.148 18.3
7/4, 8/7 8.952 20.1
3/2, 4/3 9.156 20.6
9/7, 14/9 9.360 21.1
13/10, 20/13 9.770 22.0
11/10, 20/11 12.774 28.7
5/4, 8/5 13.686 30.8
15/14, 28/15 13.891 31.3
13/9, 18/13 14.395 32.4
11/6, 12/11 17.304 38.9
11/7, 14/11 17.508 39.4
11/8, 16/11 17.985 40.5
9/8, 16/9 18.312 41.2
15/13, 26/15 18.926 42.6
15/8, 16/15 21.602 48.6
13/11, 22/13 21.901 49.3

The following tables show how 15-odd-limit intervals are represented in 27edo. Prime harmonics are in bold; inconsistent intervals are in italics.

15-odd-limit intervals in 27edo (direct approximation, even if inconsistent)
Interval and complement Error (abs, ¢) Error (rel, %)
1/1, 2/1 0.000 0.0
7/6, 12/7 0.204 0.5
15/11, 22/15 3.617 8.1
13/8, 16/13 3.917 8.8
5/3, 6/5 4.530 10.2
9/5, 10/9 4.626 10.4
7/5, 10/7 4.734 10.7
13/7, 14/13 5.035 11.3
13/12, 24/13 5.239 11.8
11/9, 18/11 8.148 18.3
7/4, 8/7 8.952 20.1
3/2, 4/3 9.156 20.6
9/7, 14/9 9.360 21.1
13/10, 20/13 9.770 22.0
11/10, 20/11 12.774 28.7
5/4, 8/5 13.686 30.8
15/14, 28/15 13.891 31.3
13/9, 18/13 14.395 32.4
11/6, 12/11 17.304 38.9
11/7, 14/11 17.508 39.4
11/8, 16/11 17.985 40.5
9/8, 16/9 18.312 41.2
15/13, 26/15 18.926 42.6
15/8, 16/15 21.602 48.6
13/11, 22/13 21.901 49.3
15-odd-limit intervals in 27edo (patent val mapping)
Interval and complement Error (abs, ¢) Error (rel, %)
1/1, 2/1 0.000 0.0
7/6, 12/7 0.204 0.5
13/8, 16/13 3.917 8.8
5/3, 6/5 4.530 10.2
9/5, 10/9 4.626 10.4
7/5, 10/7 4.734 10.7
13/7, 14/13 5.035 11.3
13/12, 24/13 5.239 11.8
7/4, 8/7 8.952 20.1
3/2, 4/3 9.156 20.6
9/7, 14/9 9.360 21.1
13/10, 20/13 9.770 22.0
5/4, 8/5 13.686 30.8
15/14, 28/15 13.891 31.3
13/9, 18/13 14.395 32.4
11/8, 16/11 17.985 40.5
9/8, 16/9 18.312 41.2
15/13, 26/15 18.926 42.6
13/11, 22/13 21.901 49.3
15/8, 16/15 22.842 51.4
11/7, 14/11 26.936 60.6
11/6, 12/11 27.141 61.1
11/10, 20/11 31.671 71.3
11/9, 18/11 36.297 81.7
15/11, 22/15 40.827 91.9

Regular temperament properties

Subgroup Comma List Mapping Optimal
8ve Stretch (¢)
Tuning Error
Absolute (¢) Relative (%)
2.3 [43 -27 [27 43]] -2.89 2.88 6.50
2.3.5 128/125, 20000/19683 [27 43 63]] -3.88 2.74 6.19
2.3.5.7 64/63, 126/125, 245/243 [27 43 63 76]] -3.70 2.39 5.40
2.3.5.7.13 64/63, 91/90, 126/125, 169/168 [27 43 63 76 100]] -3.18 2.39 5.39
2.3.5.7.13.19 64/63, 76/75, 91/90, 126/125, 169/168 [27 43 63 76 100 115]] -3.18 2.18 4.92

27et (27eg val) is lower in relative error than any previous equal temperaments in the 13-, 17-, and 19-limit. The next equal temperaments doing better in those subgroups are 31, 31, and 46, respectively.

27et is particularly strong in the 2.3.5.7.13.19 subgroup. The next equal temperament that does better in this subgroup is 53.

Rank-2 temperaments

Periods
per 8ve
Generator Temperaments MOS Scales
1 1\27 Quartonic/quarto
1 2\27 Octacot/octocat 1L_12s, 13L_1s
1 4\27 Tetracot/modus/wollemia 1L_5s, 6L_1s, 7L_6s, 7L_13s
1 5\27 Machine/kumonga 1L_4s, 5L_1s, 5L_6s, 11L_5s
1 7\27 Myna/coleto/minah/oolong 4L_3s, 4L_7s, 4L_11s, 4L_15s, 4L_19s
1 8\27 Beatles/ringo 3L_4s, 7L_3s, 10L_7s
1 10\27 Sensi/sensis 3L_2s, 3L_5s, 8L_3s, 8L_11s
1 11\27 Superpyth 5L_2s, 5L_7s, 5L_12s, 5L_17s
1 13\27 Fervor 2L_3s, 2L_5s, 2L_7s, 2L_9s, 2L_11s, etc ... 2L_23s
3 1\27 Semiaug/hemiaug
3 2\27 Augmented/Augene/ogene 3L_3s, 3L_6s, 3L_9s, 12L_3s
3 4\27 Oodako/terrain 3L_3s, 6L_3s, 6L_9s, 6L_15s
9 1\27 Terrible version of Ennealimmal
/niner
9L_9s

Commas

27edo tempers out the following commas. (Note: This assumes the val 27 43 63 76 93 100].)

Prime
Limit
Ratio[1] Monzo Cents Color name Name(s)
5 128/125 [7 0 -3 41.06 Trigu Diesis, augmented comma
5 20000/19683 [5 -9 4 27.66 Saquadyo Minimal diesis, Tetracot comma
5 78732/78125 [2 9 -7 13.40 Sepgu Medium semicomma, Sensipent comma
5 (26 digits) [1 -27 18 0.86 Satritribiyo Ennealimma
7 686/675 [1 -3 -2 3 27.99 Trizo-agugu Senga
7 64/63 [6 -2 0 -1 27.26 Ru Septimal comma, Archytas' comma, Leipziger Komma
7 50421/50000 [-4 1 -5 5 14.52 Quinzogu Trimyna
7 245/243 [0 -5 1 2 14.19 Zozoyo Sensamagic
7 126/125 [1 2 -3 1 13.79 Zotrigu Septimal semicomma, Starling comma
7 4000/3969 [5 -4 3 -2 13.47 Rurutriyo Octagar
7 1728/1715 [6 3 -1 -3 13.07 Triru-agu Orwellisma, Orwell comma
7 (12 digits) [-6 -8 2 5 1.12 Quinzo-ayoyo Wizma
7 2401/2400 [-5 -1 -2 4 0.72 Bizozogu Breedsma
7 4375/4374 [-1 -7 4 1 0.40 Zoquadyo Ragisma
7 (12 digits) [-4 6 -6 3 0.33 Trizogugu Landscape comma
11 99/98 [-1 2 0 -2 1 17.58 Loruru Mothwellsma
11 896/891 [7 -4 0 1 -1 9.69 Saluzo Pentacircle
11 385/384 [-7 -1 1 1 1 4.50 Lozoyo Keenanisma
13 91/90 [-1 -2 -1 1 0 1 19.13 Thozogu Superleap
  1. Ratios longer than 10 digits are presented by placeholders with informative hints

Music

Gene Ward Smith
Igliashon Jones
Joel Taylor
Peter Kosmorsky
Chris Vaisvil
Dustin Schallert
Brendan Byrnes

See also